KVR Audio

meloco_go · Post by **meloco_go** » Sat Jun 23, 2012 7:38 am

izonin wrote:These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".

Nope, just the clean sound.

jam92189 · Post by **jam92189** » Sat Jun 23, 2012 7:47 am

meloco_go wrote:
izonin wrote:These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".
Nope, just the clean sound.

I think he means at the higher frequencies the harmonics seem to be lower on the higher harmonics but wouldn't that help give it a smoother sound?

michu · Post by **michu** » Sat Jun 23, 2012 7:51 am

izonin wrote:
andy_cytomic wrote: These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".

how would these pictures relate to any frequency muffling? please note that the input was fed with a sine waveform, in other words only the huge peak around 1 kHz is original signal, all other frequency content represents harmonic artefacts, aliasing in red colour for viewers convenience.

also. Duende apparently tries to emulate non-linearities of analogue hardware, how faithfully is a different question. The Glue, as stated by Andy a number of times, doesn't do that at all, emulating only gain reduction behaviour of hardware. ie. you cannot overdrive it no matter how hard you try (unless you have softclip activated)

izonin · Post by **izonin** » Sat Jun 23, 2012 8:07 am

michu wrote:
izonin wrote:
andy_cytomic wrote: These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".
how would these pictures relate to any frequency muffling? please note that the input was fed with a sine waveform, in other words only the huge peak around 1 kHz is original signal, all other frequency content represents harmonic artefacts, aliasing in red colour for viewers convenience.

also. Duende apparently tries to emulate non-linearities of analogue hardware, how faithfully is a different question. The Glue, as stated by Andy a number of times, doesn't do that at all, emulating only gain reduction behaviour of hardware. ie. you cannot overdrive it no matter how hard you try (unless you have softclip activated)

Well, the harmonic artifacts are present in all examples. If you listen to the Glue vs. Analog comparison on the Cytomic site, you can clearly hear how the hi-hat and ride sound more open going through the hardware (maybe that's the side effect that wasn't modeled).

meloco_go · Post by **meloco_go** » Sat Jun 23, 2012 12:49 pm

jam92189 wrote:
meloco_go wrote:
izonin wrote:These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".
Nope, just the clean sound.
I think he means at the higher frequencies the harmonics seem to be lower on the higher harmonics but wouldn't that help give it a smoother sound?

Yes, but it doesn't mean any attenuation of top-end. By that logic the ultra-clean device shouldn't be able to resproduce any top-end at all

egbert · Post by **egbert** » Sat Jun 23, 2012 1:41 pm

@ michu
Look at the slope across the top of the (green) harmonics - on some of the examples it is more or less linear downhill slope all the way to the right hand edge of the graph. On the graph for The Glue, there is a pronounced rolloff toward the right hand edge. The upper partials are comparitively suppressed - hence less top end. This is certainly less top end on the harmonic distortion BUT the process (Andy will know what it is) which results in this roll off in top end of the harmonic distortion could easily also be applying to the wet (post-compression) full bandwidth source material (eg hi hats) so that highs in the compressed source material are suppressed (as well as higher partials of the harmonic distortion).

HcDoom · Post by **HcDoom** » Sat Jun 23, 2012 3:58 pm

None of the itb compressor is yet perfect, but The Glue is definately the one which is the closest to it's original counterpart. And even if there is some lose of high end, it's nothing that can't be fixed with fine eq after it.

billcarroll · Post by **billcarroll** » Sat Jun 23, 2012 4:03 pm

HcDoom wrote:None of the itb compressor is yet perfect, but The Glue is definately the one which is the closest to it's original counterpart. And even if there is some lose of high end, it's nothing that can't be fixed with fine eq after it.

Loss of high end with the Glue? For dogs and bats?

I see this loss of high end mentioned a few times. Are people really interpreting the graphs correctly?

djanthonyw · Post by **djanthonyw** » Sat Jun 23, 2012 4:07 pm

There is no loss of hi end with The Glue. It sounds like there is with the new PSP BussPressor though.

meloco_go · Post by **meloco_go** » Sat Jun 23, 2012 4:44 pm

egbert wrote:The upper partials are comparitively suppressed - hence less top end.

Wrong assumption.

This is certainly less top end on the harmonic distortion BUT the process (Andy will know what it is) which results in this roll off in top end of the harmonic distortion could easily also be applying to the wet (post-compression) full bandwidth source material (eg hi hats) so that highs in the compressed source material are suppressed (as well as higher partials of the harmonic distortion).

Maybe, just not enogh to say "hence". The rolloff of higher harmonics might be just because the nonlinearity is a bit smoother.

Funkybot's Evil Twin · Sat Jun 23, 2012 5:14 pm

meloco_go wrote:
egbert wrote:The upper partials are comparitively suppressed - hence less top end.
Wrong assumption.

Load up The Glue in Christian Budde's VST Plugin Analyzer then start playing around with the frequency response analysis: The Glue remains flat no matter what you throw at it. Not sure where all this talk of a loss of high end suddenly started, but there's a measurable difference between being flat, having a loss of high end, and having a hyped top end IMO.

izonin · Post by **izonin** » Sat Jun 23, 2012 5:30 pm

Funkybot's Evil Twin wrote:
meloco_go wrote:
egbert wrote:The upper partials are comparitively suppressed - hence less top end.
Wrong assumption.
Load up The Glue in Christian Budde's VST Plugin Analyzer then start playing around with the frequency response analysis: The Glue remains flat no matter what you throw at it. Not sure where all this talk of a loss of high end suddenly started, but there's a measurable difference between being flat, having a loss of high end, and having a hyped top end IMO.

I got the impression that there is loss in the high frequencies, listening to the comparisons with the hardware on the Cytomic's site. Probably the analog compressor enhances/vitalizes the high end.

egbert · Post by **egbert** » Sat Jun 23, 2012 5:34 pm

Read the quote from andy_cytomic - the developer of The Glue. What do you think he is saying when he says:

These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".

I am not pushing my own thesis, I am trying to follow his - my guess is that he knows what he is talking about.

zvenx · Post by **zvenx** » Sat Jun 23, 2012 5:55 pm

Andy didn't say that....... someone quoted his photos, then added that comment of their own.
rsp

izonin · Post by **izonin** » Sat Jun 23, 2012 5:57 pm

egbert wrote:Read the quote from andy_cytomic - the developer of The Glue. What do you think he is saying when he says:

These pictures explain the clean sound, but also the slightly muffled high frequencies. Hi-hats and rides lose some of the openness when "glued".
I am not pushing my own thesis, I am trying to follow his - my guess is that he knows what he is talking about.

No, that was my interpretation of the images. SPAN shows no suppression of the high frequencies. But the hi-hats still don't sound as open/transparent as on the hardware examples.

Cytomic 'The Glue' Compressor